Forecasting the Possibility of Agricultural Drought using Markov Chain Based on Satellite data and Standard Precipitation Index

Kolahdouzan, Mohammad Ehsan | 2013

677 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 45671 (09)
  4. University: Sharif University of Technology
  5. Department: Civil Engineering
  6. Advisor(s): Abrishamchi, Ahmad
  7. Abstract:
  8. In comparison with flood, drought is a phenomenon that happens gradually and causes significant changes in water resources, environmental and economic damages specially farming. Due to random features of involving factors in occurrence and intensity of drought, it is assumed as a random process. Drought forecast has an important role in water resources management and control. Data validation, in drought forecast, is a matter of utmost importance. In recent years, satellite data is a type of data that has been used. Markov chain has been used here because one of the features of this method is that occurrence of each variable in current stage is dependent on its previous stage. On the other hand, drought in each period could be greatly affected by previous period. Hence, drought forecast using Markov chain would have desirable results. In this research at the EGHLID city located in FARS state, satellite data of vegetation with Sixteen day time step and Two hundred and fifty meter accuracy has been converted in to monthly NDVI for each cell in each picture in GIS software; and then based on monthly data of precipitation for Twenty-five stations, SPI were calculated for each cell using IDW interpolation. At the end, the correlation between the aforementioned indexes was calculated and vegetation drought index for the future obtained using Markov chain. Results indicated that the best forecasts were in middle and end of the growing season and also the largest difference amount between real and forecasted values was at the beginning of the growing and harvest season¬¬¬¬¬
  9. Keywords:
  10. Correlation Coefficient ; Markov Chain ; Normalized Difference Vegetation Index ; Standard Precipitation Index (SPI) ; Agricurtural Drought ; Doroodzan Dam Basin

 Digital Object List